Chronic pain is a significant problem for at least half of all persons receiving opioid agonist therapy (OAT) for opioid use disorder ? i.e., buprenorphine/ naloxone (BUP) or methadone maintenance treatment (MMT). Pharmacotherapeutic pain management is challenging in OAT patients, and other behavioral options for managing chronic pain are needed. Hatha yoga may be a useful adjunctive approach for decreasing pain- related disability and pain severity, and preventing opioid misuse during OAT. There is evidence supporting its efficacy in other chronic pain populations, and yoga may target cravings and other risk factors for opioid relapse. With NCCIH R34 funding, we conducted a pilot RCT of a 12-week hatha yoga class vs. a health education class (HE; a control condition) for people with chronic pain receiving MMT (n=20) or BUP (n=20) for opioid use disorder. The primary target of the yoga intervention was decreased life interference due to pain. In this study, we measured multiple indices of feasibility and acceptability of the manualized interventions. Although we met a priori benchmarks on many of these feasibility indices, we did find that, consistent with literature on behavioral interventions in OAT patients, some participants attended only very few classes. Thus, as a next step in this research, the purpose of the current project is to develop an optimized yoga intervention package that includes key components that increase yoga dosage received, defined as time in class (a function of number of classes attended) plus home yoga practice time. We plan to use the Multiphase Optimization STrategy (MOST) to do this. We will evaluate four intervention components, all of which will be added to our standard group yoga classes, to determine their impact on total yoga dosage received. These components are: a) two added 1:1 meetings with the yoga teachers; b) provision of study teacher-led home practice videos; c) monetary incentives for class attendance; and d) text messages designed to increase intrinsic motivation for yoga practice. As a secondary goal, in keeping with the U01 program announcement, we will also demonstrate that we can conduct this research in two sites distinct from each other and from our previous site. In this study, after conducting pilot work (Phase 1), we plan to conduct a fully-powered factorial experiment that will allow us to evaluate the impact of each of the 4 intervention components on yoga dosage received. All participants will receive the core yoga intervention, with random assignment to receive or not receive each of the four intervention components outlined above. Results from Phase 2 will allow us to choose an efficient combination of intervention components that, together with standard yoga classes, maximizes yoga dosage. We will also be able to examine mechanisms by which intervention components are hypothesized to work. This project will prepare us for the next step in this research, i.e., a fully-powered, multi-site RCT testing the optimized yoga intervention (vs. a control condition), with a primary outcome of reduced pain interference.